static-site-cloner

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (medium risk: 0.30). The skill’s required workflow uses web_fetch(url="target_url", extractMode="markdown") to retrieve and ingest HTML/markdown from a user-supplied target website URL, meaning outsider-authored free text can be fetched and analyzed at runtime.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 0.90). The skill calls web_fetch(url="target_url") at runtime (SKILL.md:16-20) which will fetch arbitrary external HTML provided by the user and that fetched content is directly used to drive the agent's reproduction instructions, so the external URL "target_url" is a runtime dependency controlling prompts.